CAPITULO SEGUNDO
D.- Prescripción de la sanción disciplinaria
Participants and Setting
The participants for the study were drawn from public middle school math and science teachers located in a southeastern state of the United States. The schools from which the teachers were sampled are members of a regional education agency made up of 18 school districts in the area. County A has approximately 9,793 students, County B has approximately 10,975 students, County C has approximately 11,079 students, and County D has approximately 12,164 students. County A has 2 middle schools with a total of 21 math teachers and 21 science teachers, County B has 4 middle schools with 28 math teachers and 27 science teachers, County C has 3 middle schools with 29 math teachers and 27 science teachers, and County D has 4 middle schools with 28 math teachers and 27 science teachers.
A convenience sample of 93 middle school math and 43 middle school science teachers were chosen for this study. Once permission was obtained from each school district’s
Superintendent, an email was sent to each math teacher and each science teacher at the middle schools in the district. The email explained the purpose of the study and how the information would be collected from those who opted to participate. All respondents were selected to be part of the sample unless the teacher taught both math and science; those teachers were excluded from the sample. The number of participants sampled will exceed the required minimum for a medium effect size. According to Gall et al. (2007), 56 teachers in each group is the required minimum for a medium effect size with statistical power of .7 at the .05 alpha level meaning the total sample size needed is 112 teachers. The make-up of the sample teacher population is shown in the tables below. Table 1 and Table 2 display the gender and race/ethnicity of the sample population, Table 3 displays the grade level and subject area taught by the sample teachers, and Table 4 displays the highest degree earned by the teachers in the sample
population. Table 1 Sample Gender Gender Sample Male 30 Female 106 Table 2 Sample Race/Ethnicity Race/Ethnicity Sample White 114 Black/African American 22 Table 3
Grade Level and Subject Area Taught
Grade Math Teachers Science Teachers Total
6 27 8 35 7 23 22 45 8 6,7,8 22 21 13 0 35 21 Table 4
Highest Degree Earned
Degree Sample
Bachelor’s Degree 39
Master’s Degree 52
Educational Specialist 33
Math teachers consisted of 27 6th grade teachers, 32 7th grade teachers, and 22 8th grade teachers. There were 21 teachers who teach all three of these grades (6th, 7th, and 8th). The make- up of the math teachers by gender and race/ethnicity is displayed in Table 5 and Table 6. The highest degree earned by the math teachers is displayed in table 7.
Table 5
Math Teacher Gender
Gender Sample
Male 25
Female 68
Table 6
Math Teacher Race/Ethnicity
Race/Ethnicity Sample
White 85
Black/African American 8
Table 7
Highest Degree Earned by Math Teachers
Degree Sample
Bachelor’s Degree 15
Master’s Degree 36
Educational Specialist 30
Doctorate 12
Science teachers consisted of 8 6th grade teachers, 22 7th grade teachers, and 13 8th grade teachers. The make-up of the science teachers by gender and race/ethnicity is displayed in Table 8 and Table 9. The highest degree earned by the science teachers is displayed in table 10.
Table 8
Science Teacher Gender
Gender Sample
Male 5
Female 38
Table 9
Science Teacher Race/Ethnicity
Race/Ethnicity Sample
White 29
Black/African American 14
Table 10
Highest Degree Earned by Science Teachers
Degree Sample
Bachelor’s Degree 24
Master’s Degree 16
Educational Specialist 3
Doctorate 00
This study took place in the spring semester of the 2018-2019 school year. Math teachers in this study teach students enrolled in grades 6, 7, and 8 and follow the Georgia Standards of Excellence. The Georgia Standards of Excellence for middle school math place an emphasis on representation, problem solving, reasoning, connections, and communication focusing on
Number Sense, Expressions and Equations, Geometry, and Statistics and Probability (Georgia Standards of Excellence, 2016). Science teachers in this study teach students enrolled in grades 6, 7, and 8 and teach using the Georgia Standards of Excellence. Middle school science in
Georgia is divided by grade level, with 6th grade teaching Earth Science, 7th grade teaching Life Science, and 8th grade teaching Physical Science (Georgia Standards of Excellence, 2016). Table 11 and Table 12 display the race/ethnicity and student subgroup data for County A, Table 13 and Table 14 display the race/ethnicity and student subgroup data for County B, Table 15 and Table 16 display the race/ethnicity and student subgroup data for County C, and Table 17 and Table 18 display the race/ethnicity and student subgroup data for County D.
Table 11
Race/Ethnicity for County A
Race/Ethnicity Number of Students
American Indian/Alaskan 49 Asian/Pacific Islander 153 Black 2,287 Hispanic 688 Multi-Racial 650 White 5.966
Table 12
Student Subgroup Data for County A
Subgroup Number of Students
Male 5,012
Female 4,781
Economically Disadvantaged 5,166
Not Economically Disadvantaged 4,627
Students With Disability 1,253
Students Without Disability 8,540
Table 13
Race/Ethnicity for County B
Race/Ethnicity Number of Students
American Indian/Alaskan 26 Asian/Pacific Islander 181 Black 4,269 Hispanic 659 Multi-Racial 388 White 5,452
Table 14
Student Subgroup Data for County B
Subgroup Number of Students
Male 5,658
Female 5,371
Economically Disadvantaged 7,139
Not Economically Disadvantaged 3,836
Students With Disability 1,547
Students Without Disability 9,428
Table 15
Race/Ethnicity for County C
Race/Ethnicity Number of Students
American Indian/Alaskan 33 Asian/Pacific Islander 217 Black 5,703 Hispanic 1,368 Multi-Racial 861 White 2,897
Table 16
Student Subgroup Data for County C
Subgroup Number of Students
Male 5,618
Female 5,461
Economically Disadvantaged 7,235
Not Economically Disadvantaged 3,844
Students With Disability 1,368
Students Without Disability 9,711
Table 17
Race/Ethnicity for County D
Race/Ethnicity Number of Students
American Indian/Alaskan 10 Asian/Pacific Islander 115 Black 1,879 Hispanic 751 Multi-Racial 547 White 8,862
Table 18
Student Subgroup Data for County D
Subgroup Number of Students
Male 6,237
Female 5,927
Economically Disadvantaged 5,137
Not Economically Disadvantaged 7,027
Students With Disability 2,157
Students Without Disability 10,007
Instrumentation
The Teacher Efficacy and Attitudes toward STEM (T-STEM) survey was used to measure teacher efficacy and beliefs (Friday Institute of Educational Innovation, 2012). The purpose of the T-STEM survey was to gather information on “how confident teachers are about teaching STEM-related content, 21st century skills, and technology use in the classroom” (Friday Institute for Educational Innovation, 2012). This instrument was developed by the Friday Institute for Educational Innovation (2012) along with North Carolina State University as part of the Maximizing the Impact of STEM Outreach Project. When developing the T-STEM survey the Friday Institute of Educational Innovation used information from the Science Teaching Efficacy Belief Instrument (Riggs & Enoch, 1990), the Student Technology Needs Assessment (SERVE Center, 2005), the Student Learning Conditions Survey (Friday Institute for Educational
Innovation, 2011), and the North Carolina Department of Public Instruction professional standards (2012). Bennett (2016) used the Teaching Efficacy and Beliefs and Outcome
Expectancy Beliefs constructs of the T-Stem survey to study teacher sense of self-efficacy with regard to teaching integrated STEM in the elementary school. Bennett (2016) compared the responses of teachers at a Title I school to those at a non-Title I school. While other studies on teacher perceptions toward STEM found there was teacher interest to teach STEM but lack of time and training, along with the traditional separation of subjects into specific disciplines in middle and high school, impeded effectively implementing an integrated STEM program (Coppola, Madariaga, & Schnedeker, 2015; Ruggirello & Balcerzak, 2013).
The T-STEM consists of three validated forms, one form for elementary teachers, one form for math teachers, and one form for science teachers. The math and science teacher surveys, which are identical with only the specific subject area referenced in the survey items changing, were used (Friday Institute of Educational Innovation, 2012). A comparison chart was created by the researcher as further evidence of the analogous nature of the two surveys. See Appendix A for the comparison chart.
For this study, the T-STEM was given electronically using Google forms and was anticipated to take approximately 15 minutes to complete. While the entire T-STEM survey consists of seven subscales, only the subscales of Personal Teaching Efficacy and Beliefs (PTEB), Teaching Outcomes Expectancy Beliefs (TOEB), and STEM Instruction were used as each subscale has been independently assessed for validity and reliability. All statements were evaluated using a 5 point Likert scale where 1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, and 5 = strongly agree. Table 19 displays a breakdown of each of these subscales in terms of total items and score ranges.
Table 19
T-Stem Subscales
Subscale Number of Items Total Score Range
PTEB 11 11-55
TOEB 9 9-45
STEM Instruction 14 14-70
Note. PTEB = Personal Teaching Efficacy and Beliefs; TOEB = Teaching Outcome Expectancy Beliefs
The T-STEM Survey was not validated as a composite score, only at the subscale level. The authors of the survey discussed these subscales as themes that can be compared amongst groups. In each subscale, the higher the score the stronger the teacher’s belief in that area. For example, on the TOEB the higher the score the more the teacher believes student learning is impacted by his or her actions (Friday Institute of Educational Innovation, 2012). Table 20 displays the construct reliability for each of the forms of the T-STEM survey.
Table 20
T-STEM Survey Reliability
Construct Science Math
Personal Teaching Efficacy and Beliefs
.908 .943
Teaching Outcome Expectancy Beliefs
.814 .849
Permission was given by the Friday Institute to use the survey for educational, non-commercial purposes either “as is” or modified as long as the original source is cited. See Appendix C.
Procedures
The researcher received conditional approval from the Institutional Review Board (IRB) pending documented approval from each school district in which the study was being conducted. The researcher then sent an email to the Superintendents of each of the school districts being used for the study requesting permission to conduct the survey in the district with math teachers and science teachers. The email included an explanation of the study being conducted. Once permission to conduct the survey was received via an email response from the superintendents of each district, Institutional Review Board (IRB) permission to conduct the survey was obtained. In order to keep the data collected secure and private, respondents were only identified by demographic data along with grade level and subject area taught. There was no personally identifying data collected.
Following IRB approval, an email was sent to the principal (and assistant principal(s)) of each of the middle schools explaining the purpose of the T-STEM survey along with an
explanation of the study being done. It was requested they forward the survey to math teachers and science teachers in their school. The email forwarded to teachers instructed the recipient to click on the link provided if they wished to participate in the survey. Once the recipients clicked on the link provided in the email, they were redirected to the survey cover page where the purpose of the survey was explained to the participants along with a consent statement that instructed them to click yes or no to indicate their response. Recipients who selected no were redirected to a screen that provided a thank you statement. Recipients who clicked yes were redirected to the demographics page of the instrument. Once the recipients completed the
demographics page, they clicked next and began the T-STEM survey instrument.
At the end of a two-week period, the researcher had not received the minimum number of responses needed (56 math teachers and 56 science teachers) thus a follow up email was sent to the same middle school principals. The email was identical to the first email sent with a follow up message encouraging principals to forward the survey to their math teachers and science teachers. During this process, many school districts’ email system would tag the survey as originating outside of the school district which contributed to confusion on the part of the administrators being asked to forward the survey. After an additional two-week period, the minimum number of responses had not been received and the researcher began to email the math and science teachers in each school district directly. This email was identical to those sent to the school principals and encouraged participation. Once the minimum number of responses was obtained, the data was entered in SPSS software.
Data Analysis
The study involved conducting an independent-samples t-test to derive a statistical analysis for each dependent variable to determine if there is a difference between the mean scores of middle school teachers who teach math as compared to those who teach science (independent variable) in the areas of teacher efficacy and beliefs, teacher outcome expectancy beliefs, and the use of STEM instructional practices (dependent variables). The t-test is
appropriate for determining whether the two groups differ in their mean score for each dependent variable (Gall, Gall, & Borg, 2007). Data screening was conducted prior to the analysis
regarding data inconsistencies, outliers, and normality (Green & Salkind, 2017). A box and whisker plot was used to check for outliers for the scores on each of the subscales. Normality for each of the dependent variables was examined using a Kolmogorov-Smirnov test. The
Kolmogorov-Smirnov test was appropriate given the sample size was greater than fifty. The results of the Kolmogorov-Smirnov showed the assumption of normality was not met. An assumption of homogeneity of variance was examined using Levene’s Test for Equality of Variances (Laerd Statistics, 2015) and this was also found to be violated.
The data screening process revealed the data did not meet the assumption of normality and the homogeneity of variance was also violated. Thus the researcher decided to use a Mann- Whitney U test. The Mann-Whitney U test is an alternative to the independent samples t-test and is recommended when the data is not normally distributed and/or violates the homogeneity variance (Green & Salkind, 2011; Laerd Statistic, 2015).
CHAPTER FOUR: FINDINGS